Abstracts

Post-processing of EEG and Neuroimaging in Epilepsy Surgery Evaluation: A Head-to-head Comparison Study

Abstract number : 1.257
Submission category : 5. Neuro Imaging / 5B. Functional Imaging
Year : 2023
Submission ID : 291
Source : www.aesnet.org
Presentation date : 12/2/2023 12:00:00 AM
Published date :

Authors :
Presenting Author: Evy Cleeren, PhD – UZ Leuven

Pietro Mattioli, Department of Neuroscience, Rehabilitation, Ophthalmology, Genetics, Maternal and Child Health (DINOGMI) - University of Genoa, Genoa, Italy; IRCCS Ospedale Policlinico San Martino, Genoa, Italy – University of Genoa; Tom Theys, Department of Neurosurgery, University hospitals Leuven, Belgium; Department of Neurosciences, Research Group Experimental Neurosurgery and Neuroanatomy, KU Leuven, Belgium – UZ Leuven; Katrien Jansen, Department of Pediatric Neurology, University Hospital Leuven, Leuven, Belgium; Department of development and regeneration, KU Leuven, Belgium – UZ Leuven; Wim Van Paesschen, Department of Neurology, University hospitals Leuven, Belgium; Laboratory for Epilepsy Research, KU Leuven, Belgium – UZ Leuven; Sándor Beniczky, Danish Epilepsy Center, Dianalund, Denmark; Department of Clinical Neurophysiology, Aarhus University Hospital, and Department of Clinical Medicine, Aarhus University, Aarhus, Denmark – Danish Epilepsy Center

Rationale:
The goal of this study was to compare interictal (ii) and ictal (ic) electric source imaging (ESI) with advanced post-processing techniques of conventional neuroimaging methods.



Methods:
We included 30 consecutive patients (16M/14F, median age: 19.5y; range: 4-61y) who underwent an MRI, interictal FDG-PET, ictal SPECT, 5 day, 25-channel video-EEG monitoring and received epilepsy surgery with at least one year follow-up. This study was approved by the local ethics commission. As part of the presurgical evaluation, MRI images were post-processed using the Morphometric Analysis Program (MAP18), FDG-PET images using an anatomy-corrected asymmetry index (ACAI) and ictal SPECT images as Subtraction ictal-interictal SPECT co-registered to MRI (SISCOM). All images were discussed and interpreted during a multidisciplinary team meeting (MDTM) in which 28 patients were considered MRI positive.
ESI was performed by two analyzers using two inverse solutions (i.e., equivalent current dipole [ECD] and distributed source model [DSM]) while using an individual head model. iiESI was analyzed on a 24-hour EEG recording. icESI was analyzed on the ictal signal of the seizure in which injection for ictal SPECT was performed. An example of the multimodal workup is shown in Figure 1.
We assessed the agreement between the different ESI and imaging methods using GWET AC1 scores, using the interpretation of a modality by the MDTM on a sublobar level. To calculate accuracy measures, we used the site of resection and the one year postoperative outcome as a reference standard. A modality was considered true positive when all areas shown by the modality were resected and the patient was seizure-free; false positive when all areas were resected but the patient was not seizure-free. When not all areas shown by the modality were resected and the patient was not seizure-free, it was considered a true negative; and false negative when not all areas were resected but the patient was seizure-free. 



Results:

The concordance between the different ESI and conventional imaging methods was moderate to substantial, with the highest concordance between iiDSM and ACAI and iiDSM and MRI (GWET AC1=0.80, 95%CI:0.64-0.95); and the lowest between icESI and SISCOM (0.49 (0.29-0.68) (icDSM)-0.56 (0.36-0.75) (icECD)).

icECD was the best performing ESI method, with an accuracy similar to MRI and the post-processing of PET and SPECT (accuracy(95%CI) for icECD: 0.57(0.38-0.74); MRI: 0.73(0.54-0.88); ACAI: 0.43(0.26-0.66); SISCOM: 0.43(0.26-0.66); p >0.05). ESI had a higher sensitivity but a lower specificity than the post-processing of PET and SPECT (Figure 2). 



Conclusions:

The different neuroimaging methods, by interpretation of the MDTM, are concordant. Although none of the methods separately had a perfect accuracy, ictal ESI seemed to be more informative than interictal ESI.  This study warrants for a multimodal approach of all these methods in clinical practice, weighing for the methodological and pathophysiological characteristics of each method, tailored to the individual patient.



Funding: NA

Neuro Imaging